Between academia and industry: how data is handled today
Working with data exists in two worlds—academic and industrial—each with its own logic, pace, and concept of success. Where these worlds diverge, where they intersect, and why the most interesting things emerge at their intersection—this is what lab members Anna Kartasheva, Mikhail Meshkov, and Daniil Kovalev discuss.
In data analysis, the boundary between academia and industry is becoming increasingly permeable. Businesses formulate very specific requests: how to understand what's happening within a team or organization, what factors are associated with effectiveness, where communication gaps arise, how to identify non-obvious groups, roles, and interaction patterns, how to make data-driven management decisions. But not every such request can be immediately translated into a finished analytical product.
And here, in my opinion, the role of academia is particularly important. Researchers don't simply apply an existing tool to a new data set—they ask whether the chosen metric truly measures what we want to measure, how robust the result is, and what alternative explanations exist. In business, it's natural to expect an answer and a solution; in science, a normal research outcome can also be the refutation of the original hypothesis. It's precisely this distinction that sometimes creates tension, but also makes collaboration particularly valuable.
Academy, too, cannot exist separately from real-world challenges. Industrial data and research enable us to identify problems difficult to identify through purely theoretical work, test methods in a more complex environment, and formulate new research questions. Therefore, I would speak not so much about "whether academia is responding to business needs" as about the collaborative formulation of the problem. The most interesting projects arise when business brings a real problem, and the researcher helps transform it into a question that can be properly answered using data.
This is especially true for network analysis. Many processes in organizations cannot be understood by analyzing only individual employee characteristics: it is important to see the structure of interactions between people, teams, and departments. Such methods allow us to move from individual indicators to an analysis of the entire system of relationships. And here, the combination of fundamental methodology, data from a real organizational environment, and an understanding of the management task can yield significantly more than any of these components alone.
"Business requires certainty and repeatability, while a scientific approach allows for the possibility that a hypothesis will not be confirmed; plus, the speed of work and expectations for results differ."
Anna Kartasheva's article for IQ Media: "Science and Business: Recipes for Integration. How to Build Mutually Beneficial Collaboration."
For me, working with data in academia and in industry are two worlds, or rather, two sets, that overlap but are still quite distinct. Industry often demands proven methods, adherence to strict templates, and, most importantly, short-term results. Academia, on the other hand, allows one to focus primarily on research as a process and, most importantly, to try and explore new, untested approaches, and, most importantly, to create them. Therefore, in my opinion, combining these two fields is extremely beneficial. In one, you apply and implement knowledge into the world, while in the other, you create it. And this approach helps prevent burnout in the often monotonous analytical work, because, as the great scientist I.M. Sechenov said, "The best rest is a change of activity."
Conceptually, the worlds of science and industry are autonomous social spaces, each operating according to its own rules and competing for its own specific type of capital. Pierre Bourdieu demonstrated that each field possesses relative autonomy: it independently determines what is considered valuable, who is considered successful, and what resources enable it to achieve a higher position. In academia, the most important capital is scientific reputation, publications, peer recognition, and contribution to the production of new knowledge. In industry, economic efficiency, technological solutions, the speed of innovation, and its commercial relevance are key.
However, autonomy does not mean complete isolation. On the contrary, today these fields increasingly intersect. Science requires access to data, technologies, and business resources, while industry turns to scientific knowledge as a source of innovation. It is at the intersection of these two worlds that new forms of collaboration emerge, where fundamental research is transformed into applied solutions, and practical problems become the starting point for new scientific discoveries.
Furthermore, the position of the researcher can be viewed from two perspectives. The advantage of an academic environment is that it encourages critical thinking, provides strong methodological training, and creates conditions for rigorous hypothesis testing. A good example of this is computational social science—an interdisciplinary field in which researchers master modern methods of data analysis, machine learning, and network analysis, combining them with an understanding of social processes. This training develops the ability to accurately interpret results, consider data limitations, and critically evaluate the methods used.
Industry, by contrast, is focused on speed, efficiency, and decision-making under time constraints. Here, the practical applicability of solutions, the ability to work with large volumes of data, and the ability to quickly adapt to changing business challenges are particularly valued. However, it is precisely the combination of these two approaches that creates the most sought-after data scientists.
As a result, modern researchers are no longer limited to a single field. They are able to move freely between academia and industry, adapting their knowledge and competencies to different tasks. In academia, it fosters the development of new methods and the production of scientific knowledge, while in industry, it implements these methods in real-world processes, helping to make more informed decisions. This type of mobility allows one to simultaneously contribute to the advancement of science and create practical value for society and business.
